Related Experiment Video
Updated: Dec 16, 2025
3D Modeling of the Lateral Ventricles and Histological Characterization of Periventricular Tissue in Humans and Mouse
Published on: May 19, 2015
Dynamically constructed network with error correction for accurate ventricle volume estimation
Gongning Luo1, Wei Wang1, Clara Tam2
1School of Computer Science and Technology, Harbin Institute of Technology, Harbin, 150001, China.
Insights
This study introduces a novel deep learning method for accurate automated ventricle volume estimation (AVVE) from cardiac MRI. The approach corrects estimation errors, improving clinical applicability for cardiac disease diagnosis.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Cardiology
Background:
- Automated ventricle volume estimation (AVVE) using cardiac magnetic resonance (CMR) is crucial for diagnosing cardiac diseases.
- Current AVVE methods lack error correction, leading to inaccurate ejection fraction (EF) assessment and limiting clinical use.
Purpose of the Study:
- To develop an accurate AVVE method with integrated error correction for enhanced clinical applicability.
- To address the limitations of existing AVVE techniques in CMR image analysis.
Main Methods:
- Proposed a novel dynamically constructed deep learning framework evolving a single model into a bi-model network for direct EF correlation.
- Introduced an error correction strategy using dynamically created residual nodes with stochastic configurations and EF correlation constraints.
- Formulated an end-to-end joint optimization framework for accurate ventricle volume estimation and error correction.
Main Results:
- The proposed method significantly outperforms state-of-the-art techniques on large-scale CMR datasets.
- Demonstrated effective error correction for AVVE, a novel contribution to the field.
- Achieved high accuracy in ventricle volume estimation and EF assessment.
Conclusions:
- The developed method offers a promising solution for clinically applicable AVVE on CMR images.
- This work represents the first approach to incorporate error correction in AVVE.
- The methodology shows potential for extension to other medical index estimation tasks.
Abstract:
Automated ventricle volume estimation (AVVE) on cardiac magnetic resonance (CMR) images is very important for clinical cardiac disease diagnosis. However, current AVVE methods ignore the error correction for the estimated volume. This results in clinically intolerable ventricle volume estimation error and further leads to wrong ejection fraction (EF) assessment, which significantly limits the application potential of AVVE methods. The objective of this paper is to address this problem with AVVE and further make it more clinically applicable. We proposed a dynamically constructed network to achieve accurate AVVE. First, we introduced a novel dynamically constructed deep learning framework, that evolves a single model into a bi-model volume estimation network. In this way, the EF correlation can be built directly based on the bi-model network. Second, we proposed an error correction strategy using dynamically created residual nodes, which is based on stochastic configurations with an EF correlation constraint. Finally, we formulated the proposed method into an end-to-end joint optimization framework for accurate ventricle volume estimation with effective error correction. Experiments and comparisons on large-scale cardiac magnetic resonance datasets were carried out. Results show that the proposed method outperforms state-of-the-art methods, and has good potential for clinical application. Besides, the proposed method is the first work to achieve error correction for AVVE and also has the potential to be extended to other medical index estimation tasks.
